To represent the meaning of a word, most models use external language resources, such as text corpora, to derive the distributional properties of word usage. In this study, we propose that internal language models, that are more closely aligned to the mental representations of words, can be used to derive new theoretical questions regarding the structure of the mental lexicon. A comparison with internal models also puts into perspective a number of assumptions underlying recently proposed distributional text-based models could provide important insights into cognitive science, including linguistics and artificial intelligence. We focus on word-embedding models which have been proposed to learn aspects of word meaning in a manner similar to humans and contrast them with internal language models derived from a new extensive data set of word associations. An evaluation using relatedness judgments shows that internal language models consistently outperform current state-of-the art text-based external language models. This suggests alternative approaches to represent word meaning using properties that aren't encoded in text.

Most distributional lexico-semantic models derive their representations based on external language resources such as text corpora. In this study, we propose that internal language models, that are more closely aligned to the mental representations of words could provide important insights into cognitive science, including linguistics. Doing so allows us to reflect upon theoretical questions regarding the structure of the mental lexicon, and also puts into perspective a number of assumptions underlying recently proposed distributional text-based models. In particular, we focus on word-embedding models which have been proposed to learn aspects of word meaning in a manner similar to humans. These are contrasted with internal language models derived from a new extensive data set of word associations. Using relatedness and similarity judgments we evaluate these models and find that the word-association-based internal language models consistently outperform current state-of-the art text-based external language models, often with a large margin. These results are not just a performance improvement; they also have implications for our understanding of how distributional knowledge is used by people.

Languages vary widely in many ways, including their canonical word order. A basic aspect of the observed variation is the fact that some word orders are much more common than others. Although this regularity has been recognized for some time, it has not been well-explained. In this paper we offer an information-theoretic explanation for the observed word-order distribution across languages, based on the concept of Uniform Information Density (UID). We suggest that object-first languages are particularly disfavored because they are highly non-optimal if the goal is to distribute information content approximately evenly throughout a sentence, and that the rest of the observed word-order distribution is at least partially explainable in terms of UID. We support our theoretical analysis with data from child-directed speech and experimental work.

This paper outlines a hierarchical Bayesian model for human category learning that learns both the organization of objects into categories, and the context in which this knowledge should be applied. The model is fit to multiple data sets, and provides a parsimonious method for describing how humans learn context specific conceptual representations.

One of the central problems in cognitive science is determining the mental representations that underlie human inferences. Solutions to this problem often rely on the analysis of subjective similarity judgments, on the assumption that recognizing likenesses between people, objects, and events is crucial to everyday inference. One such solution is provided by the additive clustering model, which is widely used to infer the features of a set of stimuli from their similarities, on the assumption that similarity is a weighted linear function of common features. Existing approaches for implementing additive clustering often lack a complete framework for statistical inference, particularly with respect to choosing the number of features. To address these problems, this article develops a fully Bayesian formulation of the additive clustering model, using methods from nonparametric Bayesian statistics to allow the number of features to vary. We use this to explore several approaches to parameter estimation, showing that the nonparametric Bayesian approach provides a straightforward way to obtain estimates of both the number of features and their importance.

We present a computational Bayesian approach for Wiener diffusion models, which are prominent accounts of response time distributions in decision-making. We first develop a general closed-form analytic approximation to the response time distributions for one-dimensional diffusion processes, and derive the required Wiener diffusion as a special case. We use this result to undertake Bayesian modeling of benchmark data, using posterior sampling to draw inferences about the interesting psychological parameters. With the aid of the benchmark data, we show the Bayesian account has several advantages, including dealing naturally with the parameter variation needed to account for some key features of the data, and providing quantitative measures to guide decisions about model construction.

The additive clustering model is widely used to infer the features of a set of stimuli from their similarities, on the assumption that similarity is a weighted linear function of common features. This paper develops a fully Bayesian formulation of the additive clustering model, using methods from nonparametric Bayesian statistics to allow the number of features to vary. We use this to explore several approaches to parameter estimation, showing that the nonparametric Bayesian approach provides a straightforward way to obtain estimates of both the number of features used in producing similarity judgments and their importance.

We apply a minimum description length-based clustering technique to the problem of partitioning a set of learning curves. The goal is to partition experimental data collected from different sources into groups of sources that are statistically the same. We solve this problem by defining statistical models for the data generating processes, then partitioning them using the normalized maximum likelihood criterion. Unlike many alternative model selection methods, this approach is optimal (in a minimax coding sense) for data of any sample size. We present an application of the method to the cognitive modeling problem of partitioning of human learning curves for different categorization tasks.

An applied problem is discussed in which two nested psychological mod-els of retention are compared using minimum description length (MDL).The standard Fisher information approximation to the normalized maximum likelihood is calculated for these two models, with the result that the full model is assigned a smaller complexity, even for moderately large samples. A geometric interpretation for this behavior is considered, along with its practical implications.

Despite the popularity of connectionist models in cognitive science, their performance can often be difficult to evaluate. Inspired by the geometric approach to statistical model selection, we introduce a conceptually similar method to examine the global behavior of a connectionist model, by counting the number and types of response patterns it can simulate. The Markov Chain Monte Carlo-based algorithm that we constructed (cid:222)nds these patterns efficiently. We demonstrate the approach using two localist network models of speech perception.

This paper develops a new representational model of similarity data that combines continuous dimensions with discrete features. An al- gorithm capable of learning these representations is described, and a Bayesian model selection approach for choosing the appropriate number of dimensions and features is developed. The approach is demonstrated on a classic data set that considers the similarities between the numbers 0 through 9.

A transformational approach to visual perception is presented, in which image structure is encoded in the parameters of those transformations that produce an output maximally symmetric with the current input. Results are presented for a computer program, dubbed SMART (Symmetry Maximizing Array using Random Transformations). SMART consists of a parallel array of independent symmetry detectors. Each detector attempts to find a transformation that maximizes the symmetry between the original and the transformed configuration. The weighted output of the detectors is collated in a connection matrix, which summarizes the image structure and provides a continuously-varying measure of relative symmetry. The program is applied to constrained and random arrays, Glass (1969) figures and the detection of hidden symmetric targets. More general implications are briefly discussed.